This application requests five years of funding to sustain and expand the Mental Health Research Network (MHRN), a consortium of 13 research centers affiliated with large integrated health systems. Two new members will join the network, increasing the diversity of health system organization and adding a larger rural population. Established data and informatics infrastructure will be maintained (with reduced levels of funding). New infrastructure development will aim to address the opportunities and challenges described above and to address specific goals described in RFA MH-14-110: * Dissemination of MHRN tools and resources to the broader mental health research community * Facilitating new research collaborations with investigators outside of MHRN member institutions * Improving capacity for ongoing surveillance of mental health treatment patterns and outcomes MHRN will be governed by a steering committee of participating investigators and representatives from NIMH. An Administrative Core will house specific infrastructure activities, including: * The Informatics Unit will maintain and expand data infrastructure to support multi-site research. * The Organizational Unit will continue work to streamline administrative and regulatory processes. * The Outreach and External Collaboration Unit will continue engagement with health system partners and external stakeholders while expanding outreach to external investigators. Four specific research projects are proposed: * Reducing cardiovascular risk in adults with SMI using EMR-based clinical decision support - This cluster randomized trial will test an informatics-based intervention to reduce risk factors for cardiovascular disease among people living with severe mental illness. * Maximizing biospecimen collection from children with mental health conditions - This pilot study will evaluate strategies for collecting biospecimens from families affected by early-onset mental disorders. * Next-generation assessment using mobile devices - This pilot study will evaluate the feasibility, acceptability, and potential utility of direct assessment of behavior and neuropsychological performance (including NIMH RDoC constructs) for prediction or early detection of depression treatment response. * Automated outreach for depression treatment dropout - This pilot study will evaluate the feasibility and acceptability of population-based outreach to address early dropout from depression treatment.